One of the central results in Einstein's theory of Brownian motion is that the mean square displacement of a randomly moving Brownian particle scales linearly with time. Over the past few decades sophisticated experiments and data collection in numerous biological, physical and financial systems have revealed anomalous sub-diffusion in which the mean square displacement grows slower than linearly with time. A major theoretical challenge has been to derive the appropriate evolution equation for the probability density function of sub-diffusion taking into account further complications from force fields and reactions. Here we present a derivation of the generalised master equation for an ensemble of particles undergoing reactions whilst being subject to an external force field. From this general equation we show reductions to a range of well known special cases, including the fractional reaction diffusion equation and the fractional Fokker-Planck equation.
Continuous Time Random Walks with Reactions and Forcingwas derived in [12][13][14]. This was extended to non-linear reactions in [15]. An alternative derivation with pure death processes was given earlier in [16].The CTRW model for sub-diffusion exhibits non-ergodic behaviour. The ensemble average mean square displacement scales as a sub-linear power law in time whereas time average mean square displacements scale approximately linearly with time, with different diffusion coefficients for different realisations [17,18]. An alternate model for sub-diffusion is fractional Brownian motion (fBm). This model has a Gaussian probability density function with a time dependent diffusion coefficient and it exhibits ergodic behaviour [19]. In this article we have confined our attention to the CTRW model, however tests to distinguish between different models for sub-diffusive behaviour are an active area of research [18,20].Among the experimental systems that exhibit diffusion, reactions and forcing, the motility of populations of cells is especially interesting because of the confounding aspect of chemotaxis, where particles experience a force in proportion to a chemical gradient. The governing equations to model sub-diffusion with linear reactions and chemotaxis have been derived in [21]. Sub-diffusion with non-linear reactions and chemotaxis was considered in [22].Fractional Fokker-Planck equations and fractional reaction-diffusion equations have become increasingly important because of the growing recognition that anomalous diffusion is ubiquitous [23]. Anomalous sub-diffusion is widespread in biological systems due to traps and crowding effects. In such systems the traps may be structural traps [24, 30] or binding spots [26]. Anomalous sub-diffusion has been measured in protein movements in cells [27], in movement of lipid granules in yeast cells [28] and movements of molecules in spiny neuronal dendrites [29, 30]. The consideration of the possible anomalous diffusion of ions in nerve cells prompted the development of a fractional cable equation to model the electrical...